16th Pacific Rim Real Estate Society Conference
Wellington, New Zealand
24-27 January 2010
Approaches for Calculation of Holding Costs in the Context of
Greenfield Residential Development
Garner GO (Garner, Gary Owen)
School of Urban Development
Faculty of Built Environment & Engineering
Queensland University of Technology
[email protected]
+61 7 3138 5327 phone
+61 7 3138 1170 fax
Abstract – There are increasing indications that the contribution of holding costs and its
impact on housing affordability is very significant. Their importance and perceived high level
impact can be gauged from considering the unprecedented level of attention policy makers
have given them recently. This may be evidenced by the embedding of specific strategies to
address burgeoning holding costs (and particularly those cost savings associated with
streamlining regulatory assessment) within statutory instruments such as the Queensland
Housing Affordability Strategy, and the South East Queensland Regional Plan. However,
several key issues require further investigation. Firstly, the computation and methodology
behind the calculation of holding costs varies widely. In fact, it is not only variable, but in
some instances completely ignored. Secondly, some ambiguity exists in terms of the
inclusion of various elements of holding costs and assessment of their relative contribution.
Perhaps this may in part be explained by their nature: such costs are not always immediately
apparent. They are not as visible as more tangible cost items associated with greenfield
development such as regulatory fees, government taxes, acquisition costs, selling fees,
commissions and others. Holding costs are also more difficult to evaluate since for the most
part they must be ultimately assessed over time in an ever-changing environment based on
their strong relationship with opportunity cost which is in turn dependant, inter alia, upon
prevailing inflation and / or interest rates. This paper seeks to provide a more detailed
investigation of those elements related to holding costs, and in so doing determine the size of
their impact specifically on the end user. It extends research in this area clarifying the extent
to which holding costs impact housing affordability. Geographical diversity indicated by the
considerable variation between various planning instruments and the length of regulatory
assessment periods suggests further research should adopt a case study approach in order to
test the relevance of theoretical modelling conducted.
Keywords - Holding cost, housing affordability; regulatory assessment; opportunity cost;
Greenfield
P a g e |2
Introduction
The economic evaluation of land development projects, like many other kinds of projects, is
typically undertaken by using different measures of value based on discounted cash flows.
Therefore, the element of time is a critical determinant of viability since the discount applied
to any project is always based on discount over time. As pointed out in a recent Urbis report
(Walker et al., 2008), like all industries, time is of the essence to the land development
business. Since time is critical, it is apparent that if a project takes longer to come to
realisation, for any reason, then the costs of that project will increase. In the case of a
property development project, costs relating to that portion of time when a project is held up
are generally regarded as “holding costs”.
Holding costs can take many forms, but they inevitably involve the computation of “carrying
costs” of an initial outlay that has yet to fully realise its ultimate yield. Although sometimes
considered a “hidden” cost, it is submitted that holding costs prospectively represent a major
determinate of value. Considered in the context of housing affordability, it is therefore
potentially pervasive.
This paper focuses on the varying approaches and methodologies adopted when the
calculation of holding costs is undertaken, focussing on greenfield development. Whilst
acknowledging there may be some consistency in embracing first principles relating to
holding cost theory, a review of the literature reveals considerable lack of uniformity in this
regard. There is even less clarity in quantitative determination, especially in Australia where
there has been only limited empirical analysis undertaken. This is despite a growing quantum
of research undertaken in relation to various elements connected with housing affordability.
The end result has been a modicum of qualitative commentary relating to holding costs.
There have been few attempts at finer-tuned analysis that exposes a quantified level of
holding cost calculated with underlying rigour. It may therefore not be too surprising that this
matter has become highly prioritised in the Australian Government’s housing research
agenda.
Methods
This paper further develops previous modelling attempts seeking to quantify the impact of
holding costs on housing affordability - in particular, the consequences of extended
regulatory assessment periods as a component of holding costs. However, the focus here is to
examine holding cost theory, and how it is applied (or in some cases not applied) in the
computation of land development projects.
Since the cost of housing impacts housing affordability, examination of linkages with holding
costs therefore has potential to provide greater exposition of housing affordability equations.
Understanding the nature and composition of holding costs applying in residential property
markets (particularly greenfield development) provides a basis for understanding the impact
of indirect regulatory costs – in particular, costs which may be associated with the length of
the regulatory assessment period.
The literature review covers the background to holding cost theory and generally accepted
“first principles”. It proceeds to provide further insight by providing a comparative
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e |3
methodology of the application of holding cost theory, in the process examining linkages
with land supply problems, timeliness of regulatory assessment, and ensuing apparent
financial impacts and other perspectives.
The focus of this paper is therefore to examine holding cost theory based on the hypotheses
that it is a primary driver of housing affordability. Connections between regulatory
assessment, holding costs and housing affordability are central themes. The dimensions of
regulatory assessment as part of the “development pipeline” are given some attention, in
order to establish the extent of diversity that exists between different regulatory regimes. The
outcomes of this are compared against attempts at quantifying holding cost impact as
reviewed in the literature.
This provides additional foundation for related research, in particular the examination of
assessment periods against various holding cost elements (and / or the total quantum of
holding costs) via spreadsheet scenario analysis. This allows a comparison of outcomes
through the modelling of independent variables, such as interest rates (in particular), and the
passage of time. In this way, the impacts on housing affordability are ultimately clarified by
testing the impact of the major drivers of holding costs.
Geographically, there is considerable variation between various planning instruments and the
length of regulatory assessment periods. This implies the need to collect empirical evidence
based on various group relationship data - a case study approach. Whilst this paper stops
short of field testing predictive models that have potential to reliably quantify the impact of
planning delays, and other holding cost variables, it may be anticipated that such models
could be readily developed as a result of this preliminary research.
The Complexity of the Holding Cost Calculation
Quantifying holding costs and other related costs such as those associated with delays in
obtaining assessment and approvals can be complex depending on the Project and the
variables applying in particular circumstances. This complexity increases since holding costs
can occur over any or even all stages involved in a property development pipeline - from
initial strategic identification of a site, until construction completion and beyond.
Since holding costs are incurred over the total period of financial commitment by
stakeholders, they are impacted by various responses to market conditions existing and
changing over that time. In the case of a greenfield development, this includes not only
prevailing interest rates / investment alternatives that underpin the opportunity cost, but also
the period of investment commencing with property acquisition right through to time taken
for sales to be effected upon dwelling completion. This fundamentally involves the demand /
supply equation, adding even further complication since this equation must also take into
account the aspect of human nature itself. We are reminded of this in a recent study which
suggests that housing prices are “better explained in terms of human behaviour and social
changes than by mere trend analysis” (Small, 2009). Household lifecycles and behaviour are
now thought to be strongly relevant factors in relation to housing affordability: for example, a
recent AHURI report (Wood & Ong, 2009) found that precarious housing affordability
circumstances are particularly evident among younger couples with dependent children, a
stage in the life cycle that is associated with pressing spending needs.
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e |4
Holding cost measurement – a derivation of the EOQ model
The basic EOQ (Economic Order Quantity) model identifies the penalty associated with
ordering either too much or too little. Holding costs are in reality simply a derivation of the
EOQ model, where the shape of the “holding cost curve” demonstrates the sensitivity of the
basic EOQ model to lot-size errors when holding costs are assumed to be a strictly increasing
(though not necessarily linear) function of average inventory (Brown et al., 1986). The
premise is that the penalty associated with ordering either too much or too little is a function
not only of the size of the error but of the shape of the holding-cost curve as well.
Derivations of the EOQ model may be found in a variety of applications. For example, most
models of inventory control utilise modified versions of the EOQ formula, with the capital
cost of holding inventory able to be calculated by adding a fixed interest rate, r, times the
purchase price, C, to the out-of pocket holding cost However, this assumes the per unit
purchase price is constant, therefore where the purchase price t varies over time, methods for
computing an adjusted interest rate, r, are suggested along with modifications of well-known
heuristics and formulas for lot-sizing, with r being estimated as the sum of the unadjusted
interest rate and the average expected purchase price decrease, measured over a period
between 1/3 and 2/3 of the length of the order cycle (Berling, 2007). Other variations of the
economic order quantity (EOQ) model such as Ferguson’s (Ferguson et al., 2007) enable its
use in the case of perishable goods, such as milk, and produce. This is achieved by
considering cumulative holding cost as a nonlinear function of time. In this instance the
holding cost curve parameters can be estimated via a regression approach from the product’s
usual holding cost (storage plus capital costs), lifetime, and markdown policy. Thus, a
significant improvement in cost vis-à-vis the classic EOQ model is provided.
Some commentators determine that the holding cost rate represent outcomes of a net present
value approach, and an average cost approach, which are approximately equivalent. This has
been the approach undertaken for more complex inventory holding cost measurement. An
example of this may be seen in the measurement of inventory in a two-product system
involving joint manufacturing and remanufacturing (Çorbacıoğlua & van der Laan, 2007)
whom conclude that the correct holding cost rates deviate from traditional valuation
methodology, with impact on operational performance demonstrable. Nonetheless, it is the
EOQ model that forms the basis for examining the cost of holding money.
The Relevance of Opportunity Cost & the Use of Capitalisation and
Discounting
The Present Value & Discount Factor
The holding cost of an investment is generally regarded as being equivalent to opportunity
cost. Opportunity cost has been, in its simplest form, described as a term used by economists
to depict the scenario when someone forgoes one opportunity to take another (Powell &
Stringham, 2004). Another definition (Miles et al., 2004) describes opportunity cost as being
interest that could have been earned that is forgone: this forgone interest represents the
opportunity cost associated with receiving a dollar in the future rather than today.
Consequently, today’s value, or the present value, of the dollar to be received in a given time
period should be reduced by the cost of the “lost opportunity” over that same time period.
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e |5
The concept of opportunity cost therefore involves the calculation of a present value, on the
basis that we are solving for the difference between the current day value of a compounded
future amount. The amount of interest that could have been earned during the term of an
investment – the compound interest – represents the difference between the present value and
the future value amount, and is known as the discount. Guthrie describes the discount as
being the “shrinkage” that occurs when an amount of money is moved back in time at the
compound interest rate (Guthrie & Lemon, 2004). This is also more generally known as the
opportunity cost, or perhaps more colloquially, opportunity “lost”.
The general present value formula is expressed as:
FV
PV = (1 + i ) n
Where
PV
FV
i
n
is the Present Value
is the Future value
s the interest rate per period
s the total interest periods
+ i )− n is typically expressed since it is easier
that way for the algebraic calculator. The factor (1 + i )− n is the discount factor (also known
The transposed formula PV
= FV
(1
as the present worth of 1 factor), that is simply the reciprocal of the accumulation factor, i.e.
(1 + i )n which is the basic tool for solving accrued compound interest.
Thus, we can determine that the discount factor for an investment that can earn 9.5% per
annum over 15 years is (1+0.095)-15. Thus, an asset worth $100,000 in 15 years time can be
calculated to have a present value of $25,632. The difference between the asset’s future worth
of $100,000 and the present value, i.e. in this case $74,368, represents the “opportunity cost”
of investing $25,632 over 15 years, or the amount of interest that could have been earned at
the relevant compound interest rate, had it been invested. Therefore we have a formulae for
Opportunity Cost oC as:
oC = FV − ⎡ FV (1 + i )− n ⎤
⎢⎣
⎥⎦
It is this imputed value over time that is fundamental to the concept of “holding cost”. If an
investment is made in a certain asset that requires it to be held during a period in which incurs
no growth, then the amount of interest foregone because of the need to “hold” the investment
is equivalent to the “opportunity cost” of holding the asset. In other words, one depiction is
that it represents the interest foregone due to the expense made on the outlay.
Selection of Interest Rate Applicable for the Calculation of Opportunity Cost
Obviously, the longer the time taken, the greater the cost of holding the asset. However, what
is often the greatest difficulty to determine is the selection of the interest rate. As pointed out
(Darnell & Evans, 1988), the rate of interest provides the correct measure only if the relevant
alternative to holding cash balances is holding interest bearing assets. That suggests that the
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e |6
opportunity cost measurement should reflect the utility that is anticipated to having to forgo
as a result of making the choice to hold money. The definition given for “Opportunity cost”
therefore relies upon a comparison between holding non-interest bearing money, and the best
alternative providing the greatest financial yield.
The usual approach to measuring the cost of holding money is to note that by holding cash
balances an individual foregoes income that could be earned on an interest-bearing asset
(Darnell & Evans, 1988). From this, Darnell states, it is usually inferred that the 'opportunity
cost' of holding cash is determined by the rate of interest. Further, any debate has been over
the selection of a data proxy for the rate of interest (e.g. should it be a short/long rate? the
dividend price ratio? the whole structure of interest rates? etc.). The value v of holding noninterest bearing money is zero, since the future value of $1 remains $1, no matter the passage
of time: the face value remains the same. In that instance, v1 = 1 . In the case of holding
interest bearing money the formula is equivalent to the impact of r the nominal interest rate is
v 2 = (1 + r ) . However, as Darnell argues, the value of holding a physical good is equivalent
to a change in value due to η inflation, expressed as v3 = (1 + η ) . Thus, the results for each
possibility can be expressed in the following table 1:
Table 1- Derivation of financial gains foregone (the "best alternatives" for holding cash)
Action
Holding non-interest
bearing money
Holding non-interest
bearing money
Relevant alternative
action
Holding interest bearing
money
Percentage gain foregone
Holding a physical good
(v3 − v1 ) / v1 = η
(v 2 − v1 ) / v1 = r
Taken from The Holding Cost of Money (Darnell & Evans, 1988)
This argues that in determining the cost of holding these money balances is the greater of the
nominal interest rate, and the inflation rate. This is because whilst the monetary gain foregone
in the case of purchase of an interest bearing asset is the nominal interest rate, the monetary
gain foregone in the case of a good is the rate of inflation. This identifies the potential gain
foregone willingly, in order to enjoy the benefits of holding the asset. Accordingly, the
general formula for the expected cost of holding money may be expressed as1:
oC = max (r ,η )
Variability Caused by Period of Holding & Other Timing Factors
Reed suggests that, in relation to a property asset, the calculation for measuring the cost of
the holding period (or property “reversion”) is either the application of capitalisation rate to
an income stream (if the property is income producing), or conducting a discounted cash flow
1
A number of interesting points are noted (Darnell & Evans, 1988) whom state that (1) the real rate of interest is never the holding cost of
non-interest-bearing money. The real rate of interest may be seen as the opportunity cost of buying a good when holding an interest bearing
asset is perceived as the best alternative. (2) In studies of hyperinflation, the opportunity cost off holding real balances has been identified as
the expected rate of inflation. Since in such episodes the inflation rate persistently exceeds the nominal rate of interest, the analysis
presented provides the explicit theoretical justification for this practice.
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e |7
analysis (DCF) if there is an irregular steam of inflow and / or outflow payments (Reed,
2007). The latter computes the present value of an expected reversion, and in the case of a
property model the income stream and reversion are valued in one operation.
Regardless, the longer the holding period, the greater the risk, and therefore the greater the
discount rate used in such analysis. Reed states that this applies equally for leveraged or nonleveraged investments since there is an amortised cost in the former, or otherwise an
opportunity cost acquired in the latter case. This is in general agreement with the Adams
explanation of present value and time (Adams et al., 1968) whom states that in an effective
market, the price of land will reflect capitalisation of the anticipated future flow of net rent.
Until the time of development, the capitalisation process suggests a time path for land prices.
A distinguishing feature of vacant land, however, is that up to the time it is developed the
return to the owner is zero, or if we consider taxes and related expenses, negative.
Theoretically, then, if the development of the land has been anticipated, the price of vacant
land should tend to follow a time path determined by the discounting of its value at
development at the prevailing interest rate. Changes in expectations, interest rates and
holding costs, market imperfections, and short term construction requirements will lead to
divergence of prices from the path. Relationships between land prices and relevant variables
from the economy are to be anticipated. If we assume V at the time of development t, V is
itself the present value of an expected series of net returns, and an appropriate rate of
discount, i, the present value P, assuming continuous discounting, is as follows (Adams et al.,
1968):
P = V / e it
Thus the relative rate of change of the present value, with respect to t is as follows:
dP / dt = −i
P
Where
r
V
t
P
i
or − ⎛⎜⎝ i + r ⎞⎟⎠
is the rate of real estate taxation
is the value (at the time of development)
is the time of development
is the present value
is the appropriate rate of discount
In other words, the price of an undeveloped piece of land can be expected to grow at the rate
(i + r ) where i corresponds to the net rate of return which can be earned on other comparable
investments. Adams points out that in a perfectly operating market, the present values of
properties will be aligned to their anticipated values to the expected dates at which the
properties will be developed. If the factors which determine development value and date of
development are taken into account, undeveloped land prices may be expected to increase
over time at the rate ( i + r). This is entirely the result of capitalisation and discounting.
Overall, the costs of housing may relate to construction costs, land costs, costs of land
purchase and eventual sale (i.e. taxation and professional fees), developers profit for risktaking, and also financial costs including interest costs and opportunity costs. However, it is
the latter that is considered here, and includes (Eccles et al., 1999):
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e |8
•
•
•
the prevailing level of interest rates;
the length of time that the development takes to complete;
the length of time that the development takes to produce income or sell.
As a minimum, holding costs will relate to at least the rate applicable to the funding of a
development project, according to the nature of the Project. The generally accepted principle
is that the development moneys will be outstanding for an average of half the period during
which the estate is being developed and sold. Assuming a two year life the interest allowance
is calculated on the development costs including the contingency allowance (Whipple, 1995).
Whipple, in evaluating cash flow analysis, rightly emphasises the importance of timing on the
profitability of development projects. Static models ignore such a sensibly conceived scenario
analysis.
Taxation & Liquidity Effects
Other factors might also be included under the general ambit of “holding costs”. For example,
land taxes may not be neutral in their economic impacts due to liquidity effects. Liquidity
effects of land taxes may be in the form of holding cost effects or capitalization effects
(Bourassa, 1992). Bourassa also recognises that “holding cost” effects may occur when land
is being withheld from development for non-financial reasons, such as the direct benefits of
land ownership. Such non-financial reasons might also include processing delays by
approving bodies and other planning matters that impact on time. Capitalization effects may
occur when there are imperfections in capital markets which prevent the acquisition of land
for otherwise viable projects.
This augurs well with earlier work completed (Bourassa, 1988) which examines the liquidity
effect results from increases in the rate applied to land. The incentive effect is due simply to
the increase in supply that occurs as the excise effect of the tax is reduced. The liquidity
effect has two components. One is the effect on current landowners, who must bear increased
holding costs and who are thereby encouraged to improve their properties or sell to someone
who will. The other component is the obverse of increased holding costs and is due to
capitalization of the tax in land value. Reduced land values make it easier for potential
developers to acquire land. Bourassa concludes that the effect on current landowners, who
must bear holding costs in the form of land taxes, is encouragement to improve their
properties to maximize return on investment - or sell to someone who will do so. The other
component of the liquidity effect is simply the obverse of increased holding costs;
nonetheless economists generally agreeing that increases in taxes on land result in decreases
in land value.
Impact of Land Supply
Another perspective is the extent of house price volatility due to restriction, or otherwise, of
land supply by governments. Commonly referred to as “land banking behaviour”, this
strategy impacts not only the behaviour of property developers, but also housing prices – and
therefore, affordability. Constraints of planning decisions also impact the supply equation.
Such constraints have been described to typically include transport, infrastructure,
environmental impact, competing land uses, and construction capacity (Tse, 1998). However,
these constraints are not applied uniformly and an argument exists that the amount of
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e |9
available land, and the supply of housing, may at time relate to political considerations
outside of what might be otherwise justified by analysing population and household growth.
This leads Tse to conclude that not only land supply, but also planning controls, development
processes and marketing practices are important determinants of housing supply. It may
therefore be demonstrated that land banking behaviour is inevitably governed by general
economic conditions. Furthermore, in uncertain economic conditions, there may be greater
uncertainty about future housing price appreciation which could actually have a negative
effect upon the land-holding costs: i.e. uncertainty increases the expected future value of the
vacant land.
In examining these issues, Tse calculates an equation that long-term land holding costs
should cover interest costs on the basis that the amount of land sales by the government and
land in developers’ land banks tend to decrease when market interest rates increase. The
conclusion reached here is that the rate of interest can be viewed as a kind of land-holding
cost, since a developer’s optimal amount of land bank occurs when the expected marginal
rate of return of land holdings equals the rate of interest. This has been expressed (Tse, 1998)
as follows:
θ ⎛⎜⎝ A ⎞⎟⎠ − rL s.t. A > L
max k
=
L, A
A− L ,
Where:
k
rate of return
L
loan amount
A
amount of land in land bank
θ ( A ) expected return from holding (A) amount of land in land bank
r
interest rate to finance land holdings
Thus, the maximisation of the rate of return on equity is a result of choosing both the amount
of land in a land bank, and the amount of loan.
Holding costs may also at times work somewhat in reverse to what would normally be
expected. For example, market fluctuations may also impact on the viability of lot releases
resulting in an amended staged release or holding back of lots until a positive return can be
realistically anticipated (Walker et al., 2008). An evolutionary economics approach to the
analysis of land and property markets suggests that, at least in the short term, policies that
impose extra costs on developers, especially at a time of relatively static prices, may lead to
reduced development output (David, 2008). In this instance, the opportunity “cost” of holding
may become an opportunistic gain; however this ignores risk since holding lots longer prior
to release may not always produce a positive result.
The Treatment of Holding Costs by Commercially Available Models
Holding cost computations by commercial available software are typically provided in two
ways. Using Estate master as an example (Development Feasibility, and Development
Management Modules) the following may be observed:
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e | 10
Firstly, the “holding period” is obviated by assumptions contained within discounting
calculations in the DCF analyses / feasibility. This is the period in which an investment is
intended to be held based on investor expectations - from the point of initial financial
commitment (acquisition), to the sales revenue period. By default, it takes into account
factors such as anticipated market growth and revenue span. Interest on borrowings (and
interest received on re-investment of surplus funds) is incorporated in the discount rate.
Financing charges including interest on outlays is included by default and represent part of
the total development cost. Interest earned on deposit in a trust account (often utilised in an
acquisition transaction) is computed over the time that deposit sits in the trust account –
however the interest is divided evenly between the seller (Land Owner) and the buyer
(Developer)2: in some circumstances this may prove too prescriptive. Furthermore, these
holding costs, although computed, are not separately identifiable in the project summaries.
There may also be an argument that rare hyperinflationary conditions cannot be taken into
account, at least from a holding cost point of view, i.e. where inflation exceeds the interest
rate. The problem compounds where such conditions are not reflected in revenues received
from property.
Secondly, there is a separate, readily identifiable input category denoted “Land Holding
Costs” which is in fact a repository for capital expenditure line item or items representing
financial commitments during construction. Whilst these items can be escalated (or left as a
fixed cost), this component is provided for the inclusion of items such as insurance, council
rates and land taxes and the like, incurred during the time of property development – and
entered as whole dollar amount. These are not operational costs, but “once-off” or relatively
infrequent capital items incurred by a developer during the development phase – so they do
represent bona fide holding costs. In theory, land acquisition cost could also be included here;
however this is more appropriately sited under its own section where the payment regimes
and settlement details can be detailed, along with other acquisition cost items such as
valuation charges and legal expenses. The opportunity cost of these items, i.e. interest
equivalent incurred over time based on their capital outlay, is picked up - but imputed into the
cash flow itself and not separated out for later identification.
It may be concluded that the identification of holding costs, although generally incorporated
in commercially available development models, are not readily identifiable. This is despite
the separation of identifiable “land holding cost” capital line items.
Methodological Deficiency Despite Acknowledgement of the
Significance of Holding Costs by State Government
The Queensland Government’s recent ‘Affordable Housing Strategy’ (QHAS) acknowledges
holding costs due to costs associated with delays in obtaining assessment and approvals can add
up to $20,000 per dwelling to the end price (Queensland Housing Affordability Strategy, 2007).
These are denoted as being “development holding costs during the assessment period” Even
though the QHAS does not elucidate their computation methodology, some believe this to be a
conservative figure, and highlight the extent to which these costs can escalate. As an example,
an RDC Media Release (Elliott, 2007) calculated that in a recent Queensland development
project the tax and regulatory charges accounted for 26% of the purchase price of $579,000. It
was pointed out that excessive delays and massive court costs (on appeals) all result in
2
Both the deposit percentage and interest on deposit are optional inputs
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e | 11
excessive holding costs. In this example, the interest bill alone on the holding cost associated
with delays in council assessment was calculated to be $8,928. However, the analyst provides
limited information as to either how this cost was derived, or any detail on the methodology
used. It also ignores other holding costs associated over the total development timeframe; for
example, opportunity costs commencing with commitment upon land acquisition, re-financing
requirement, if any, and financial commitments during construction.
Operating in tandem with the QHAS in Queensland has been the newly established Urban
Land Development Authority. Its mandate reflects QHAS philosophy, in particular that
related to housing affordability - and specifically, the speeding up of property development
“red tape” processes. The ULDA intends to make housing more affordable by addressing
factors that they perceive impacts on the price of new housing. This is stated ("Urban Land
Development Authority website," 2009) to include “getting land to market faster,
streamlining development approvals, and simplifying planning requirements.” The primary
way this is intended to be achieved is by speeding up the development assessment process.
However, whilst the ULDA appears to agree with QHAS conclusions by stating that “delays
in the development assessment process can increase development holding costs between
$15,000 to $20,000 per dwelling, which is typically passed on to the end purchaser”, again
there is no indication of methodology used to derive this amount.
Lack of Clarity in Identifying Regulatory Assessment as a
Component of Holding Costs in the Greenfield Residential
Development Pipeline
The quantum of time taken by regulatory authorities to assess and consider applications for a
particular development represents part of the holding cost calculation, and may even be
demonstrated to represent a major component. However, these costs are not always well
informed or clarified even though they are repeatedly acknowledged as significantly
impacting housing affordability. It has been recently observed (National Housing Supply
Council - State of Supply Report, 2009) that the relationship between housing costs and
planning regulations, charges and procedural requirements has been raised regularly in the
course of inquiries into housing affordability. However, many of these inquiries3 struggle to
quantify, or even identify, various holding cost components.
Although research is emerging in these areas – notably through AHURI (Gurran et al., 2008)
- there have been only limited attempts to quantify the relative weight or proportionate cost of
planning related costs to a development project which might ultimately determine impact on
housing affordability. The best estimate provided in the AHURI positioning paper supports
assertions by the sector that taxes, levies and compliance costs (including costs of meeting
planning regulations and holding costs) now amount to about a third of the cost of new house
and land packages. However, Gurran notes the methodology for deriving this figure at “just
under $30,000 to the price of a block of land” is unclear4.
3
The NHSC report indicates examples such as the Department of Community Services and Health, National Housing Strategy, AGPS,
Canberra, 1991 and, more recently, the Productivity Commission (Productivity Commission, First home ownership) and the Senate Select
Committee on Housing Affordability in Australia (Senate Select Committee on Housing Affordability in Australia, A good house is hard to
find), June 2008.
4
It is noted that the final report from Gurran has yet to issue, with the next stage - empirical research - needed to verify the range of costs
(including holding costs), under differing geographical scenarios. It is also noted that the case study design for the empirical research phase,
includes provision for the calculation of both “time” and “holding cost” against a generic fee schedule containing each process cost, building
or development control requirement, and other planning related costs or charges.
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e | 12
The Residential Development Council (RDC) acknowledges the time cost of excessive delays
in gaining development approval is a significant cost with significant blow-outs in the
timeframe to process applications (Reasons to be fearful? Government taxes, charges and
compliance costs and their impact on housing affordability. Summary Report - Residential
Development Costs Benchmarking Study, March 2006., 2006). However, whilst observing
that holding costs (interest costs, rates, land tax etc) increase in line with the amount of time
it takes to prepare and assess development approvals, the calculation methodology is not
transparent. Somewhat paradoxically, the report states that “these costs have previously been
hidden from discussions on housing affordability”.
In Queensland, the Development Application process form part of the “Residential
Development Pipeline”. It is a large component of the regulatory assessment process.
Consideration of the elements of this may therefore assist in understanding the generic stages
and the expected timeframes likely to be encountered regardless of geographical location.
The elements within this “pipeline” are summarised in Figure 1 below and further detailed at
Table 2 appended to this report.
Figure 1- Simplified Generic Property Development Pipeline. Adapted by the author from sources modelled
by Qld & Federal Australian Governments & Eccles (Barker, 2008; Eccles et al., 1999; National Housing Supply Council
- State of Supply Report, 2009)
A typical total development timeframe lasting somewhere from six to sixteen years as
indicated by the above graphic, translates to a typical holding cost period from between four
to twelve years. These periods will inevitably be site specific, with the holding cost period
relating to a point of commencement aligned with initial investment commitment (occurring
somewhere between stages 1 and 2), and concluding upon sale realisation for the whole
investment (occurring somewhere between stages 6 and 8). Note in the above graphic (Figure
1) the Holding cost period is to an extent indeterminate at the extremities (thus the timeline
bar fades at either end); this can only be fully determined on a site by site basis.
These time variations alone, superimposed by interest rate variations over the time period and
the time required for full realisation, all contribute to the difficulty in arriving at a rigorously
computed holding cost calculation(s). Tranched financial arrangements for land acquisition,
re-financing during the course of a land development project (typically undertaken especially
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e | 13
in the case of larger projects), and various market constraints additional to those mentioned,
all add further complexity.
Conclusions
This paper has examined various models utilised for both defining and measuring holding
costs. Whilst most ultimately rely upon derivations of the Present Value / discounting
approach, the application of these “first principles” varies widely. As a result, the
methodology used in calculating holding costs also has wide variation. On many occasions,
the methodology utilised is not readily apparent, including disclosure of major assumptive
variables such as interest rate(s) and timing. This lack of information makes it difficult to
determine the degree of rigour that has been applied, thus does does not provide confidence
in the derived outcomes. Even commercially available applications, whilst incorporating
holding cost calculations within their models, do not fully disclose these costs as a separately
identifiable item.
In some instances, holding costs are even completely ignored in determining the total costs
involved in the development pipeline. Difficulties in their calculation are typically due to
uncertain or imprecise timelines, as well as the additional complexity of holding cost
methodology, liquidity effects and other aspects. Whilst a generic development pipeline
model can be considered, it is apparent that wide variations exist in the nature of holding
costs which have great dependency on site specific variables. This complexity in deriving
holding cost calculation may therefore explain why commentators usually provide vague or
even no detail when applying holding cost theory to support public policy, or specific land
development projects.
Despite this lack of detail, significant resources have been poured into policies designed to
specifically inhibit the holding cost effect in Australia as part of addressing the broader issue
of housing affordability. In the case of Queensland, this includes the implementation of the
Queensland Housing Affordability Strategy, and the creation of the Urban Land Development
Authority.
Whilst recognising that holding costs are only one contributor to the housing affordability
equation, there needs to be significantly more research into its underlying nature and effects,
and in particular an analysis over time. The need for a broadly based analysis by regions and
towns in Australia, i.e. empirical case study analysis, cross-referencing with a rigorous
international comparison study, is indicated. Additional consideration of further market and
non-market variables and their likely impact on housing affordability would also be required
in order to assist in determining the total impact of holding costs.
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e | 14
Table 2- The Generic Greenfield Property Development Pipeline & The Impact of Time
Stage
(commencing from
perceived Demand
Increase)
Milestone
Time
(6 years minimum, to
16 years maximum)
Detail
PIFU Residential
Development Pipeline
nomenclature
Typical Holding Cost
period incurred by
developer from initial
investment or
commitment
(4 years min. to 12
years max.)
1. Strategic
2. Gazettal of
identification rezoning/ material
and
change of use
designation
of new land
release area
DEVELOPER
IDENTIFIES VIABLE
SITE
Regulatory Constraints
operate, e.g. planning,
building consents, site
acquisition / purchase;
other constraints
2–4 years
1–3 years
Identification
of master
planned area
(in Qld,
within defined
Urban
Footprint)
• Broadhect
are Land
Rezoning under
local government
planning
instruments is
generally initiated
by the proponent –
time dependant on
scale and
complexity
3. Negotiation of
infrastructure levies
and detailed structure
planning
4. Statutory subdivision
and development
approval
5. Major civil works,
servicing of allotments and
issue of new titles
6. Land Sale
8. Dwelling
AND / OR
Completion
7. Development
approvals and
dwelling
construction
DEVELOPMENT IS CONSTRUCTED
Additional finance restructuring typically undertaken
DEVELOPER RAISES MONEY
Market constraints impact timeframes which vary
considerably (e.g. interest rates, bankers / investor
attitudes, land bought forward)
1–3 years
6 months – 2 years
1–2 years
Landowner/ developer
undertakes the
development/ structure
planning process with a
view to obtaining the
necessary approvals –
time usually depends on
the quantum of
government
departments responsible
Issue of statutory
development/subdivision
approvals is the
responsibility of the
relevant local authority
which responds to
developer-initiated
applications (road layouts,
lot sizes and dimensions)
generally on a stage-bystage basis
• Lot Approval
Completion and certification
of the construction works
(undertaken by the
landowner/developer) by
approval agencies subdivisions usually
constructed in stages of
around 50 lots - development
of a large subdivision may
therefore occur over a number
of years.
• Operational Works
• Lot production
• Lot registration
6 months -2
Years
OR
9–12 months
housing design,
approval and
construction may be
undertaken by a
lot purchaser or
by a
developer/builder
who intends to
offer a house and
land package
• Dwelling
approval
milestone
Adapted by the author from sources modelled by Qld & Federal Australian Governments & Eccles (Barker, 2008; Eccles et al., 1999; National Housing Supply Council - State of Supply Report, 2009; "Urban Land
Development Authority website," 2009)
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e | 15
References
Adams, F. G., Milgram, G., Green, E. W., & Mansfield, C. (1968). Undeveloped Land Prices
During Urbanisation - A Micro-Empirical Study Over Time. Review of Economics &
Statistics, 50(2), 248.
Barker, R. (2008). Residential Land Supply and Dwelling Trends in the Gold Coast:
Department of Infrastructure & Planning, Queensland
Berling, P. (2007). The capital cost of holding inventory with stochastically mean-reverting
purchase price. European Journal of Operational Research
Bourassa, S. C. (1988). Pittsburgh's Urban Tax Reform. American Journal of Economics &
Sociology, 47(2), 166-166.
Bourassa, S. C. (1992). Economic effects of taxes on land: A review. American Journal of
Economics & Sociology, 51(1), 109-113.
Brown, R. M., Conine, T. E. J., & Tamarkin, M. (1986). A Note on Holding Costs and Lot
Size Errors. Decision Sciences, 17(4 ), 603-610.
Çorbacıoğlua, U., & van der Laan, E. A. (2007). Setting the holding cost rates in a twoproduct system with remanufacturing RSM Erasmus University, P.O. Box 1738,
Burg. Oudlaan 50, 3000 DR Rotterdam, The Netherlands
Darnell, A. C., & Evans, J. L. (1988). The holding cost of money Applied Economics, 20(3),
395 - 406.
David, A. (2008). Mapping out the regulatory environment and its interaction with land and
property markets: Department of Urban Studies, University of Glasgow. (Q. s. P. a. C.
o. H.-E. Ltd
Eccles, T., Sayce, S., & Smith, J. (1999). Property and Construction Economics. UK:
International Thompson Business Press.
Elliott, R. (2007). Myth #7: Developers Charge More for Inner City Apartments - Because
They Can: Property Council of Australia
Ferguson, M., Jayaraman, V., & Souza, G. C. (2007). Note: An application of the EOQ
model with nonlinear holding cost to inventory management of perishables. European
Journal of Operational Research, 180(1), 485-490.
Gurran, N., Ruming, K., Randolph, B., & Quintal, D. (2008). Planning, government charges,
and the costs of land and housing (No. ISSN: 1834-9250 ISBN: 1 921201 32 0):
Australian Housing and Urban Research Institute - UNSW-UWS Research
CentreSydney Research Centre
Guthrie, G. L., & Lemon, L. D. (2004). Mathematics of Interest Rates and Finance: Pearson
Prentice-Hall.
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010
P a g e | 16
Miles, E., Berens, G., & Weiss, M. A. (2004). Real Estate Development - Principles and
Process (Third Edition ed.): Urban Land Institute.
National Housing Supply Council - State of Supply Report. (2009). National Housing Supply
Council - Department of Families, Housing, Community Services and Indigenous
Affairs
Powell, B., & Stringham, E. (2004). Housing Supply & Affordability: Do Affordable Housing
Mandates Work? USA: Reason Public Policy Institute
Queensland Housing Affordability Strategy. (2007). Queensland Government - Office of
Urban Management, Department of Infrastructure
Reasons to be fearful? Government taxes, charges and compliance costs and their impact on
housing affordability. Summary Report - Residential Development Costs
Benchmarking Study, March 2006. (2006). Property Council of Australia - Residential
Development Council / UrbisJHD
Reed, R. G. (Ed.). (2007). The Valuation of Real Estate (The Australian Edition of the
Appraisal of Real Esate 12th Edition ed.). Canberra: Australian Property Institute.
Small, G. (2009). Household dynamics and housing affordability: a more human approach to
understanding the direction of house prices Australia and New Zealand Property
Journal, 2(2), 81-87.
Tse, R. Y. C. (1998). Housing Price, Land Supply and Revenue from Land Sales. Urban
Studies, 35(8), 1377-1392.
Urban Land Development Authority website. (2009, 15 October 2009). from
http://www.ulda.qld.gov.au/
Walker, R., Courtney, M., Shing, C., & Robertson, J. (2008). Residential Land Development
Study Prepared for the UDIA (Qld) by URBIS Brisbane Qld Job Code BPE0163;
Report Number B081B)
Whipple, R. (1995). Property Valuation & Analysis (1st ed.). Sydney NSW: The Law Book
Co Ltd.
Wood, G., & Ong, R. (2009). The dynamics of housing affordability: movements in and out of
housing affordability stress 2001–2006 (No. ISSN: 1834-7223 ISBN: 978-1-92161013-4): Australian Housing and Urban Research Institute - RMIT Research Centre
Western Australia Research Centre
16th Pacific Rim Real Estate Society Conference, Wellington, New Zealand. 24-27 January 2010